Dynamic Buffer Management Using Per-Queue Thresholds

IEEE Journal on Selected Areas in Communications, 1999

Recently, there has been much interest in using active queue management in routers in order to protect users from connections that are not very responsive to congestion notification. A recent Internet draft recommends schemes based on random early detection for achieving these goals, to the extent that it is possible, in a system without "per-flow" state. However, a "stateless" system with first-in/first-out (FIFO) queueing is very much handicapped in the degree to which flow isolation and fairness can be achieved. Starting with the observation that a "stateless" system is but one extreme in a spectrum of design choices and that per-flow queueing for a large number of flows is possible, we present active queue management mechanisms that are tailored to provide a high degree of isolation and fairness for TCP connections in a gigabit IP router using per-flow queueing. We show that IP flow state in a router can be bounded if the scheduling discipline used has finite memory, and we investigate the performance implications of different buffer management strategies in such a system. We show that merely using perflow scheduling is not sufficient to achieve effective isolation and fairness, and it must be combined with appropriate buffer management strategies.

Seamless Interconnection for Universal Services. Global Telecommunications Conference. GLOBECOM'99. (Cat. No.99CH37042)

An objective of the next. generation network is the accommodation of services wit,h different QoS requirements. This indicates that the network should provide special mechanisms in order to prioritize the access to network node resources, such as link capacity and buffer space. We studied the performance of sharing buffer space and link capacity bet.ween sessions, the traffic of which is modeled by independent. general Markov-Modulated Fluid Process (MMFP) sources. For scheduling we use the Generalized Processor Sharing (G P S) policy, and improve previous upper bounds on the queue occupancy distributions. As an example of combining G P S with buffer management we apply our results to complete buffer sharing with virtual partitioning (VP+GPS). We also derive results on the resource allocation trade-off, with applications in traffic management and admission control. This work was supported in part by the NY State Center for Advanced Technologies in Telecommunications (C ATT) , and Sumitomo Electric Industries (USA). The work was done when G. Lapiotis was a research fellow in the NY State CATT, Polytechnic University.

IJCSNS, 2006

This paper investigates the problem of enabling Quality of Service (QoS) of multimedia traffic at the input port of high-performance input-queued packet switches using a simulation-based evaluation. We focus on the possibility of assuring QoS of multimedia traffic in such switches by implementing traffic prioritization at the input port where each input-queue has been modified to provide a separate buffer for each of the service classes. The multimedia traffic can be categorized into three classes based on its real-time properties and loss tolerance, and assigned a separate queue for each class. We select appropriate models for each of three types of traffic: video, voice, and data. Then, we propose an efficient dynamic scheduling strategy by implementing multimedia traffic prioritization at the input port of input-queued packet switches. Simulation-based comparisons show that while the static priority scheme is beneficial for highest priority class at the expense of the others, the dynamic prioritization serves fairly well all the classes in terms of delay and loss requirements.

Computer Communications, 2005

This paper addresses scheduling and memory management in input queued switches having finite buffer with the objective of improving the performance in terms of throughput and average delay. Most of the prior works on scheduling related to input queued switches assume infinite buffer space. In practice, buffer space being a finite resource, special memory management scheme becomes essential. Maximum weighted matching (MWM) algorithm, which is known to be the optimal in terms of throughput for infinite buffer case turns out to be sub optimal in the presence of memory limitations. We introduce a buffer management scheme called iSMM (Integrated Scheduling and Memory Management) that can be employed jointly with any deterministic iterative scheduling algorithm. We applied iSMM over iSLIP, a popular scheduling algorithm, and study its effect under various input traffic conditions. Simulation results indicate iSMM to perform better than iSLIP and MWM both in terms of throughput and delay especially under non-uniform traffic.

IEEE Journal on Selected Areas in Communications, 1995

We address the problem of des%ning optimal buffer management policies in shared memory switches when packets already accepted in the switch can be dropped (pushed-out). Our goal is to maximize the overall throughput, or equivalently to minimize the overall loss probability in the system. For a system with two output ports, we prove that the optimal policy is of push-out with threshold type (POT). The same result holds if the optimality criterion is the weighted sum of the port loss probabilities. For this system, we also give an approximate method for the calculation of the optimal threshold, which we conjecture to be asymptotically correct. For the N-ported system, the optimal policy is not known in general, but we show that for a symmetric system (equal traf6c on all ports) it consists of always accepting arrivals when the buffer is not full, and dropping one from the longest queue to acco11Lmod8fe the new arrival when the buffer is full. Numerical results are provided which reveal an interesting and somewhat unexpected phenomenon. While the overall improvement in loss pmbability of the optimal POT policy over the optimal coordinate-convex policy is not very significant, the loss probab~ty of an individual output port remains approximately constant as the load on the other port varies and the optimal POT policy is applied, a property not shared by the optimal coordinateconvex policy.

Proceedings Third IEEE International Symposium on Object-Oriented Real-Time Distributed Computing (ISORC 2000) (Cat. No. PR00607), 2000

There have been many debates about the feasibility of providing guaranteed Quality of Service (QoS) when network traffic travels beyond the enterprise domain and into the vast unknown of the Internet. Many mechanisms have been proposed to bring QoS to TCP/IP and the Internet (RSVP, DiffServ, 802.1p). However, until these techniques and the equipment to support them become ubiquitous, most enterprises will rely on local prioritization of the traffic to obtain the best performance for mission critical and time sensitive applications. This work explores prioritizing critical TCP/IP traffic using a multi-queue buffer management strategy that becomes biased against random low priority flows and remains biased while congestion exists in the network. This biasing implies a degree of unfairness but proves to be more advantageous to the overall throughput of the network than strategies that attempt to be fair. Only two classes of services are considered where TCP connections are assigned to these classes and mapped to two underlying queues with round robin scheduling and shared memory. In addition to improving the throughput, cell losses are minimized for the class of service (queue) with the higher priority.

Computer Networks, 1999

Efficient and fair use of buffer space in an Asynchronous Transfer Mode (ATM) switch is essential to gain high throughput and low cell loss performance from the network. In this paper a shared buffer architecture associated with threshold-based virtual partition among output ports is proposed. Thresholds are updated based on traffic characteristics on each outgoing link, so as to adapt to traffic loads. The system behavior under varying traffic patterns is investigated via simulation; cell loss rate is the quality of service (QoS) measure used in this study. Our study shows that the threshold based dynamic buffer allocation scheme ensures a fair share of the buffer space even under bursty loading conditions.

2016

In this work, we study the problem of buffer management in network switches from an algorithmic perspective. In a typical switching scenario, packets with different service demands arrive at the input ports of the switch and are stored in buffers (queues) of limited capacity. Thereafter, they are transferred over the switching fabric to their corresponding output ports where they join other queues. Finally, packets are transmitted out of the switch through its outgoing links to their next destinations in the network. Due to limitations in the link bandwidth and buffer capacities, buffers may experience events of overflow and thus it becomes inevitable to drop some packets. In other switching models, packets that are sensitive to delay are dropped if they exceed a specific deadline inside the queue. We consider multiple models of switching with the goal of maximizing the throughput of the switch. If all packets are treated equally, i.e., corresponding to the besteffort concept of the...

2008

Grid technologies are emerging as the next generation of distributed computing, allowing the aggregation of resources that are geographically distributed across different locations. The network remains an important requirement for any Grid application, as entities involved in a Grid system (such as users, services, and data) need to communicate with each other over a network. The performance of the network must therefore be considered when carrying out tasks such as scheduling, migration or monitoring of jobs. Network buffers management policies affect the network performance, as they can lead to poor latencies (if buffers become too large), but also leading to a lot of packet droppings and low utilization of links, when trying to keep a low buffer size. Therefore, network buffers management policies should be considered when simulating a real Grid system. In this paper, we introduce network buffers management policies into the GridSim simulation toolkit. Our framework allows new policies to be implemented easily, thus enabling researchers to create more realistic network models. Fields which will harness our work are scheduling, or QoS provision. We present a comprehensive description of the overall design and a use case scenario demonstrating the conditions of links varied over time. * Corresponding author. Tel.: +34 967 59 92 00 ext. 2693; fax: +34 967 59 93 43

Lecture Notes in Computer Science

Emulation of Output Queuing (OQ) switches using Combined Input-Output Queuing (CIOQ) switches has been studied extensively in the setting where the switch buffers have unlimited capacity. In this paper we study the general setting where the OQ switch and the CIOQ switch have finite buffer capacity B ≥ 1 packets at every output. We analyze the resource requirements of CIOQ policies in terms of the required fabric speedup and the additional buffer capacity needed at the CIOQ inputs: A CIOQ policy is said to be (s, b)-valid (for OQ emulation) if a CIOQ employing this policy can emulate an OQ switch using fabric speedup s ≥ 1, and without exceeding buffer occupancy b at any input port. For the family of work-conserving scheduling algorithms, we find that whereas every greedy CIOQ policy is valid at speedup B, no CIOQ policy is valid at speedup s < 3 √ B − 2 when preemption is allowed. We also find that CCF in particular is not valid at any speedup s < B. We then introduce a CIOQ policy, CEH, that is valid at speedup s ≥ p 2(B − 1). Under CEH, the buffer occupancy at any input never exceeds 1 + j B−1 s−1 B−1 s−2 m. We also show that a greedy variant of the CCF policy is (2, B)-valid for the emulation of non-preemptive OQ algorithms with PIFO service disciplines.